the novel protein suppressed in lung cancer down … · chang-tze ricky yu, phd,*† jiun-yi hsia,...

ORIGINAL ARTICLE

The Novel Protein Suppressed in Lung Cancer Down-Regulated in Lung Cancer Tissues Retards Cell Proliferation

and Inhibits the Oncokinase Aurora-A

Chang-Tze Ricky Yu, PhD,*† Jiun-Yi Hsia, MD,‡§� Yun-Chih Hseih, PhD,¶ Li-Jen Su, PhD,#Tien-Chiang Lee, Master,* Chia-Feng Ku, Master,* Ke-Shin Chen, Master,* Jou-May Maureen Chen, Master,*

Tong-You Wade Wei, Master,* Yuan-Chii Gladys Lee, PhD,** Chi-Ying F. Huang, PhD,††Yu-Chung Wu, MD,‡‡ Chiou-Ying Yang, PhD,� and Shih-Lan Hsu, PhD¶§

Introduction: In an attempt to search for genes with abnormalexpression in cancers, Suppressed in Lung Cancer (SLAN, alsoknown as KIAA0256) is found underexpressed in human lung cancertissues by quantitative real-time PCR (Q-RT-PCR). The study setout to characterize SLAN protein and explore its cellular functions.Methods: SLAN or its specific short hairpin RNA, full length orvarious deletion mutants were overexpressed in 293T or lung cancercell lines, and cell proliferation, cell cycle, mitosis progression, andspindle configuration were surveyed.Results: SLAN and its deletion mutants are localized to manysubcellular locations such as endoplasmic reticulum (ER), nucleus,nucleolus, spindle pole and midbody, suggesting SLAN may func-tion as a multifunctional protein. Overexpression of SLAN per se orits short hairpin RNAs (shRNAs) inhibits or accelerates cell prolif-eration through prolonging or shortening mitosis. Time-lapse micro-scopic recording reveals that cells overexpressing exogenous SLANare arrested in mitosis or cannot undergo cytokinesis. SLAN 2–551mutants drastically arrest cells in mitosis, where �- and �-tubulin aredisorganized. SLAN employs C-terminal to interact with Aurora-A,a key mitosis regulator and an oncogenic kinase associated with awide range of human cancers. SLAN negatively regulates the

activity of Aurora-A by directly inhibiting kinase activity in vitro orreducing the level of active Aurora-A in cells. SLAN is frequentlyreduced in lung cancer tissues overexpressing Aurora-A, arguing forthe necessity to suppress SLAN during the Aurora-A-associatedcancer formation.Conclusions: Taken together, we have identified a novel proteinSLAN downregulated in lung caner, having multiple subcellularlocalization including spindle matrix and midbody, inhibiting cellproliferation and Aurora-A.

Key Words: SLAN, KIAA0256, Mitosis, Tumor suppressor,Aurora-A.

(J Thorac Oncol. 2011;6: 988–997)

With the accomplishment of human genome project,biomedical research has approached the postgenomic

era. It becomes imperative to prioritize studies of the geneswith fundamental importance in human genome. Cancers arenotorious for their high prevalence and continue to be one ofthe leading causes of death in the world. Cancers arise withlargely unknown mechanisms, leading to the difficulties fortheir therapeutic treatment. Growing evidence demonstratesthat cancers are caused by accumulated gene mutations,1,2

prompting us to look for genes with altered expression in thedisease.

Lung cancer, classified into two broad categories his-tologically, i.e., small cell carcinoma and non-small cellcarcinoma, is a major cause of cancer-related mortalityworldwide with more than 1 million new cases diagnosed peryear.3,4 A range of prognostic biomarkers involving genesthat regulate cell cycle progression and apoptosis, and inva-sion and metastasis have been described in lung cancer. Forexample, certain polymorphisms, expression level, and sub-cellular localization of Aurora-A, a mitosis regulator withoncogenic activity, are suggested to serve as a prognosticfactor of lung cancers.5–7 The oncogenic Aurora-A is a keyregulator of mitosis and is found overexpressed in a widerange of human cancers, prioritizing Aurora-A as a target ofanticancer therapy and thereby stimulating the investment onthe development of Aurora-A inhibitor.8–15

*Graduate Institute of Biomedicine and Biomedical Technology, †Departmentof Applied Chemistry, National Chi Nan University, Puli, Nantou; ‡Depart-ment of Surgery, Taichung Veterans General Hospital, Taichung; §Instituteof Toxicology, Chung Shan Medical University, Taichung; �Institute ofMolecular Biology, Chung Hsing University, Taichung; ¶Department ofEducation and Research, Taichung Veterans General Hospital, Taichung;#Graduate Institute of Systems Biology and Bioinformatics, National Cen-tral University, Jhongli City; **Graduate Institute of Biomedical Informat-ics, Taipei Medical University, Taipei; ††Institute of Clinical Medicine,National Yang-Ming University, Taipei and; ‡‡Division of Thoracic Sur-gery, Department of Surgery, Taipei Veterans General Hospital and Schoolof Medicine, National Yang-Ming University, Taipei, Taiwan.

Disclosure: The authors declare no conflicts of interest.Address for correspondence: Shih-Lan Hsu, PhD, or Chiou-Ying Yang, PhD,

Department of Education and Research, Taichung Veterans GeneralHospital, No. 160, Section 3, Chung-Gang Road, Taichung 407, Taiwan.E-mail: [email protected]

Chang-Tze Ricky Yu, PhD, and Jiun-Yi Hsia, MD, contributed equally tothis manuscript.

Copyright © 2011 by the International Association for the Study of LungCancerISSN: 1556-0864/11/0606-0988

Journal of Thoracic Oncology • Volume 6, Number 6, June 2011988

To explore more molecular anticancer targets, we hadadopted microarray analyses to search for the new genes withpotential involvement in lung cancer formation. Severalgenes are noted because of their altered expression patterns inlung cancer. From a list of such genes, suppressed in lungcancer (SLAN also known as SECISBP2L or KIAA0256) isspecifically selected because of the inhibitory activity ofSLAN on cell proliferation and Aurora-A. SLAN is a novelgene with limited reports focusing on it so far. SLAN con-tains a ribosomal L7Ae/Gadd45 signature at its C-terminalwhich is analogous to the functional C-terminal fragment ofSBP2, a protein responsible for selenocysteine insertion;however, the function of SLAN remains to be determined.16

Altered expression of SLAN is detected in schizophrenia,17

age-related cataract,18 and cancers,19,20 hinting SLAN mightbe involved in the pathogenesis of some human diseases. Ouranalyses in the studies reveal the involvement of SLAN incancer formation. SLAN is down-regulated in a number oflung cancer tissues at messenger RNA (mRNA) and proteinlevel. SLAN is localized to several subcellular locations andnegatively regulates cell proliferation through the interfer-ence with mitosis progression. At last, SLAN interacts withand inhibits Aurora-A, a mitotic oncokinase21–23 and a keymitosis or meiosis regulator,24–26 providing a connection ofSLAN to molecular oncogenesis and mitosis regulation.

MATERIALS AND METHODS

Tissue Procurement, Total RNA Preparation,and Reverse Transcription

The study protocol had the approval of the ethicscommittee at Taipei Veterans General Hospital.27 No patientshad previously received any neoadjuvant treatment such aschemotherapy before the surgery. All patients gave informedconsents and signed the consent form individually. The sam-ples were obtained from a nonnecrotic area of the tumor andfrom adjacent nontumorous tissue. Study samples, includingtumor and adjacent normal tissues, were obtained after oper-ation, and adjacent normal tissues were derived from neigh-boring site outside of the tumor. The tissue was immediatelyplaced in cryovials, frozen in liquid nitrogen, and stored at�80°C until analysis. Both tumor and adjacent nontumortissues for subsequent quantitative real-time polymerasechain reaction (Q-RT-PCR) studies were confirmed by pa-thologists. RNA preparation and analysis were performedaccording to the previous study.28 Briefly, the quality of thetotal RNA was determined using Spectra Max Plus (Molec-ular Devices/MDS Analytical Technologies, Toronto, Can-ada) and had an OD260/OD280 ratio ranging from 1.9 to 2.1.RNA was subjected to reverse transcription with randomhexamer primers. To hydrolyze contaminating DNA in theRNA preparations, RNA was incubated with amplification-grade DNase I (Life Technologies, Carlsbad, CA). Afterincubating the reaction mixture, the reaction was stopped byheating at 65°C. After DNase treatment, the RNA was sub-jected to reverse transcription reaction by the ThermoScriptRT-PCR system (Life Technologies), and complementaryDNAs (cDNAs) were then used in the Q-RT-PCR.

Quantitative Real-Time Polymerase ChainReaction

Q-RT-PCR was used to measure the mRNA expressionlevels between adjacent normal and lung tumors by using384-well plates (ABI PRISM 7900HT Sequence DetectionSystem, Applied Biosystems/Life Technologies, Carlsbad,CA). The cDNAs were served as templates (diluted 200�)for Q-RT-PCR by using TaqMan Universal PCR Master Mixkit (Applied Biosystems/Life Technologies). Each 10 ml ofQ-PCR reaction mixture contained 5 ml of 2� TaqManUniversal Master Mixture (Applied Biosystems/Life Tech-nologies), 4 ml of 200� diluted cDNA product mixture, and1 ml of PCR forward and reverse primers and probe. Tostandardize the quantification of the selected target genes,DDX5 served as internal controls27 and was quantified on thesame plate as the target genes. The cycle threshold (CT) valueof DDX5 was used to normalize the target gene expressionvalues, referred to as the �CT, which was used to adjustdifferences among samples. The reaction products analyzedby agarose gel electrophoresis showed that many of theamplifications contained no detectable PCR product. TheAssays-on-Demand IDs (Applied Biosystems/Life Technol-ogies) for SLAN and DDX5 are Hs00299746 m1 andHs00189323 m1. On the basis of paired adjacent normaland tumor samples of 24 patients, we used Wilcoxon signed-rank test to determine statistic significance.

Cell CultureThe cell lines used in this study were purchased from

American Type Culture Collection. The culture condition for293T and HeLa is Dulbecco/Vogt Modified Eagle’s MinimalEssential Medium with 5% fetal bovine serum, A549, H1299,and H460 in Roswell Park Memorial Institute 1640 with 10%fetal bovine serum, 1% nonessential amino acids, and 1%sodium pyruvate. Moreover, 2 mM glutamine, 100 U/mlpenicillin, and 100 g/ml streptomycin were added in allmedia. All cell culture-related reagents were purchased fromInvitrogen-Gibco-Life technologies (Carlsbad, CA). Cells aremaintained in a humidified incubator at 37°C in the presenceof 5% CO2.

Preparation of Cell Extracts, Western BlotAnalysis, and Immunocoprecipitation

The cell extracts were prepared by extraction buffer,which consists of 50 mM Tris pH 7.5, 0.1% sodium dodecylsulfate (SDS), 1% NP40, 0.5% sodium deoxycholate, 1%Triton X-100, 5mM ethylenediaminetetraacetic acid, 150 mMNaCl, and 150 mM KCl. Protein concentrations were deter-mined using the Bradford assay (Bio-Rad Laboratories, Her-cules, CA). Equal amounts of total lysates were used forfurther analyses or loaded onto a 10% SDS-polyacrylamidegel electrophoresis (SDS-PAGE) and then transferred onto apolyvinylidene fluoride (PVDF) membrane. The PVDF mem-brane was blocked with 5% skim milk/Tris buffered salinewith Tween 20 (TBST) (150 mM sodium chloride, 20 mMTris, 0.1% Tween-20, pH 7.6). Primary antibodies wereincubated with the membrane as the titer of 1:1000 at 4°C for2 hours. The membranes were washed with TBST at roomtemperature for 10 minutes and repeated for three times.

Journal of Thoracic Oncology • Volume 6, Number 6, June 2011 Novel Protein SLAN

Copyright © 2011 by the International Association for the Study of Lung Cancer 989

Secondary antibodies conjugated with alkaline phosphatasewere added for 1 hour at room temperature followed bywashing with TBST 30 minutes for three times. 5-Bromo-4-chloro-3-indolyl phosphate and p-Nitro-Blue tetrazoliumchloride (Zymed Laboratories, San Francisco, CA) wereadded to develop the membrane.

Indirect Immunofluorescence AnalysisCells were washed with phosphate-buffered saline

(PBS) and fixed with cold methanol at �20°C for 20 minutes.Sequentially, cells were incubated with primary antibodies atroom temperature for 1 hour followed by wash three timeswith TBST and then incubated with secondary antibody—Tetramethyl Rhodamine Iso-Thiocyanate or fluorescein iso-thiocyanate (Santa Cruz Biotechnology, Delaware Avenue,CA) and 4�,6-diamidino-2-phenylindole (DAPI; Sigma, MO)for 1 hour. After washing with TBST, the samples weremounted with 90% glycerol containing antifading reagentp-phenylenediamine. The fluorescence images on the cover-slips were analyzed using an Olympus fluorescence micro-scope (Olympus Model BX51, Japan).

Flowcytometric Analysis and Mitotic IndexBriefly, 1 � 106 cells were trypsinized, washed with

PBS, and fixed in 70% ethanol, then washed with PBS,incubated with 100 �g/ml RNase at 37°C for 30 minutes,stained with propidium iodide (50 �g/ml), and analyzed on aFACScan flow cytometer. The percentage of cells in differentphases of the cell cycle was analyzed using Cell-FIT software(Becton-Dickinson, Mountain View, CA). The mitotic indexwas measured by visualization of cells with propidium iodidestaining under fluorescent microscopy. Approximately, 200cells were counted at each time point. Cells in mitosis werejudged by the appearance of condensed chromosomes.

Construction of SLAN Expression VectorspEGFP-SLAN was obtained by PCR cloning adopting

human full-length SLAN cDNA clone purchased from mamma-lian genome collection American Type Culture Collection (no.MGC-26224) as a template. Two primers with RsrII cutting site,so called CPO site, forward (5�-GATGACGGTCCGGAC-CGAGCCCCCACGGAGC-3�) and reverse (5�-GTACACG-GACCGTTACGTAGTTTGCGTTGTGTAATTAG-3�), wereused to amply SLAN cDNA by thermocycler machines (Gene-Amp PCR System 9600, Perkin-Elmer, Waltham, MA). Thefirst amino acid of SLAN was removed by skipping the firstcode in the sequence of PCR primer, which ensures the forma-tion of enhanced yellow-green variant of green fluorescentprotein (EGFP)-SLAN fusion protein after transfection of cul-tured cells. The PCR product was then applied to RsrII digestion(New England Biolabs, MA) to create CPO cloning sites. On theother hand, pEGFP-C1 vector (Invitrogen-Gibco-Life technolo-gies) with CPO site created previously was also treated withRsrII followed by a dephosphorylation reaction to prevent self-ligation. The resulted linearized vector was then ligated to theSLAN cDNA with T4 DNA ligation enzyme (New EnglandBiolabs, MA). Each deletion construct was created by usingcorresponding restriction enzyme set, such as RsrII and HindIIIfor 2-455, and CPO and HindIII for 2-599 and then ligated to

linearized EGFP empty vector. pEYFP-endoplasmic reticulum(ER) (BD Biosciences Clontech, Mountain View, CA) encodesa fusion protein consisting of enhanced yellow fluorescent pro-tein (EYFP), the ER targeting sequence of calreticulin, which iscloned at the 5� end, and the sequence encoding the ER retrievalsequence, KDEL, which is cloned at the 3� end.

Knockdown of SLANThe SLAN shRNAs used in the studies were purchased

from Academia Sinica (Taipei, Taiwan). The sequences of theshort hairpin RNAs are listed below with underlined SLAN targetsequences: C, CCGGCCTGATTACTTGTTACCCATTCT-CGAGAATGGGTAACAAGTAATCAGGTTTTTG; D,CCGGCGATCTCTGAAGTAAATGAAACTCGAGTTTC-ATTTACTTCAGAGATCGTTTTTG; E, CCGGCGCCT-GTTTCTACAGAGTATACTCGAGTATACTCTGTAGA-AACAGGCGTTTTTG; F, CCGGCCATGAATCATGT-GGAATCATCTCGAGATGATTCCACATGATTCATGG-TTTTTG; G, CCGGGCCTCATTTGTGGAAAGTGAT-CTCGAGATCACTTTCCACAAATGAGGCTTTTTG. Thecells were transfected with the shRNAs first. After 24 to 72hours, cells were harvested for protein expression or appliedto related analyses such as cell proliferation assay or flowcytometry.

Cell Proliferation AssayCells were seeded in 24-well plates and transfected

with desired constructs tagged with EGFP. After a day, thecells were replated with low density to 10-cm dish, so thatcells could not contact each other. The cells were allowed toproliferate for 36 hours, and the percentage of cells withproliferation judged by formation of “mini-colony” with cellnumber �2 was counted. If cells could not proliferate, theyremained single in the 10-cm dish after 36-hour culture.

RESULTS

SLAN is Down-Regulated in Collected LungCancer Tissues

To analyze the expression pattern of SLAN in cancertissues, we have performed Q-RT-PCR in 24 pairs of lungcancerous and adjacent normal tissues. SLAN was founddown-regulated in a great portion of the cancer biopsies(Figure 1A). The comparison between the expression ofSLAN in pooled cancerous and pooled normal tissues re-vealed a significant p value (Figure 1B).

Motif Prediction and Subcellular Localizationof SLAN

SLAN is a protein with 1101 amino acids. The proteinsequence of SLAN is highly conserved in vertebrate (Supple-mentary Figure 1A, SDC 1, http://links.lww.com/JTO/A70),revealing the potential importance of SLAN. There is no homo-logue found in insects and lower eukaryotes. There are someputative functional motifs/domains in human SLAN obtainedfrom sequence prediction such as ER membrane retention signal(2DRAP5), nuclear localization signals (NLS; 246RRRR249,396PKRAKSQ402, and 425KKLQEALSKAAGKKNK441), PESTmotifs (252HPTAESSSEQDIDSDSGYCSPKHSNNPQAA286 and

Yu et al. Journal of Thoracic Oncology • Volume 6, Number 6, June 2011

Copyright © 2011 by the International Association for the Study of Lung Cancer990

http://links.lww.com/JTO/A70

924RDLLNSSITSTTSTLVPGMLEEEEDEDEEEEEDYTH959), asumoylation site (360KDE362), a D box (490RKPL493), a ribo-somal L7Ae Gadd45 signature (676GLREVTKHMKLNK-IKCVIISPNCEK700), a tyrosine kinase phosphorylation site(704kggldealy712), and a coiled coil domain (754GAESLF-NKLVELTEEARKAYKDMVAAMEQEQAEEALK790). Col-lectively, the motif prediction suggests SLAN might be localizedto many subcellular compartments and could be subjected to anumber of posttranslational modifications. To uncover thesubcellular localization of SLAN, indirect immunofluores-cence using anti-SLAN antibody was performed in A549cells. SLAN was detected distributed around the microtubulenetwork in interphase (Supplementary Figure 1A, SDC 1,http://links.lww.com/JTO/A70), which provides a frameworkfor the ER formation. Moreover, DsRed-SLAN was colocal-ized to where EYFP-ER marker resided (SupplementaryFigure 1C, SDC 1, http://links.lww.com/JTO/A70), in linewith the prediction that SLAN has an ER membrane retentionsignal. When cells progressed to mitosis, SLAN was foundassociated with spindle pole with amorphous but notfiber-like appearance (Supplementary Figure 1B, SDC 1,http://links.lww.com/JTO/A70). Sequentially, SLAN moved tospindle midzone and midbody in telophase and cytokinesis.

To dissect the cis elements of SLAN controlling thesubcellular localization of the protein, a series of SLAN deletionmutants were created and expressed in 293T cells. As shown inFigure 2C, SLAN full length and its truncate mutants resided atnucleus, nucleolus, nuclear matrix but not nucleolus, spindlepole, and whole cell. Careful calculation (Supplementary

Figure 1D, SDC 1, http://links.lww.com/JTO/A70) revealsthat SLAN contains two NLS, two nucleus export signals,and a nucleolus localization signal (Supplementary Figure1E, SDC 1, http://links.lww.com/JTO/A70).

SLAN is Associated with Spindle Matrix-LikeStructure

Similar to the pattern of endogenous SLAN in A549mitotic cells, EGFP-SLAN was localized to the proximity ofspindle pole in the mitosis of H1299, 293T, and HeLa (Figure2A), resembling so called spindle matrix in appearance.29

Spindle matrix is reported sensitive to detergent but notnocodazole in frog and fly,29,30 prompting us to examine thedistribution of SLAN in mitosis treated with the two reagents,respectively (Figure 2B). Incubation of cells with nocodazoledepolymerized spindle and disrupted SLAN; on the otherhand, Triton X-100 did not interfere with spindle but dis-turbed the association of SLAN with spindle matrix-likestructure, implying a novel property of the SLAN-associatedstructure.

SLAN Retards Cell Proliferation by DisturbingMitosis

The negative association of SLAN with lung cancertissues fueled us to test the effect of SLAN overexpression orknockdown on cell proliferation. Because the stable cell lineoverexpressing exogenous SLAN could not be established, thecell proliferation assay based on transient expression ofEGFP-SLAN was performed. Most of the cells overexpress-

FIGURE 1. Underexpression of SLAN in lung cancer tissues. A, The mRNA expression level of SLAN was determined by Q-RT-PCR in 24 pairs of lung cancer and adjacent normal tissues. The experiments were performed in triplicate, and the resultswere normalized against the expression level of DDX5 mRNA in each sample and plotted with white bar as tumor and blackbar as normal tissue. B, The box plot shows the data distribution as a grouping classification and indicates that there is a sta-tistically significant difference between the pooled tumor tissues and the pooled adjacent nontumor tissues by Wilcoxonsigned-rank test. SLAN, suppressed in lung cancer; Q-RT-PCR, quantitative real-time polymerase chain reaction; mRNA, mes-senger RNA.








ing EGFP-SLAN failed to proliferate in both cell lines (Fig-ures 3A, B). By contrast, SLAN shRNAs knocked downSLAN and accelerated cell proliferation (Figure 3C). More-over, knockdown of SLAN in 293T cells stimulated cellproliferation in medium containing 0.1% serum, where con-trol cells almost ceased growing (Figure 3D), indicating thatdepletion of SLAN offers cells advantage to grow in theenvironment without sufficient serum.

To dissect the mechanisms by which SLAN or SLANshRNA inhibits or stimulates cell proliferation, we examinedthe effect of exogenous SLAN or SLAN shRNA on cell cycleprogression (Figure 4A). The 293T cells harboring EGFP,EGFP-SALN, or SLAN shRNA were subjected to flowcyto-metric analysis. EGFP-SLAN arrested cells in G2/M, andhence, cell cycle could not progress to G1, leading to areduction of G1; on the contrary, knockdown of SLAN accel-erated G2/M and S, resulting the accumulation of G1. Toclosely and continuously examine the effect of SLAN on cellproliferation, the cells overexpressing EGFP or EGFP-SLANwere applied to time-lapse microscope. As expected, the cellsoverexpressing EGFP had normal mitosis progression (Figure4B, upper), which went through mitosis within 80 minutes.Intriguingly, the cells harboring EGFP-SLAN either arrestedin mitosis (Figure 4B, middle) or left mitosis with a failedcytokinesis, which in turn produced binucleated cells (Figure4B, lower). Approximately, 40% or 9% of EGFP-SLANexpressing cells were binucleated or arrested in mitosis (Fig-ures 4C, D), roughly matching the difference in the cellproliferating activity between EGFP or EGFP-SLAN trans-fected 293T cells (Figure 3B). In addition to SLAN fulllength, various SLAN deletion mutants were shown to stopmitosis with different potency. SLAN 2–551 induced thehighest mitotic arrest (Figure 4E), disturbed the localizationof �- and �-tubulin, and triggered subsequent malformationof spindle (Figure 4F).

SLAN Binds and Inactivates Aurora-ATo unravel the molecular activity of SLAN, we initi-

ated a screening for its interaction proteins. Luckily, one ofthe mitosis regulators Aurora-A interacted with C-terminal ofSLAN (Figures 5A, B), consistent with previous predictionthat there is a coiled-coil domain at SLAN C-terminal, whichfunctions for protein-protein interaction. Besides, the inter-action of Aurora-A and SLAN was independent on the kinaseactivity of Aurora-A (Figure 5C). Partial colocalization ofSLAN and Aurora-A was detected in 293T cells (Figure 5D).Interestingly, SLAN could reduce the level of active Auro-ra-A in cells (Figure 5E) and inhibited the autophosphoryla-tion of Aurora-A and Aurora-A-catalyzed phosphate incor-poration into an artificial substrate myelin basic protein in invitro kinase reactions (Figure 5F). Collectively, the SLANdown-regulated in lung cancer tissues with an inhibitoryactivity on cell proliferation is shown to inactivate an onco-genic kinase Aurora-A.

Underexpression of SLAN and Overexpressionof Aurora-A are Detected in Lung CancerTissues

Aurora-A is a well-documented oncoprotein associatedwith a wide range of human malignancies including lungcancer.5–7 On the other hand, SLAN possesses anticell pro-liferating activity and is able to inhibit Aurora-A, indicatingSLAN is epistatic to Aurora-A. Accordingly, it is logicallyassumed that cancer tissues with up-regulated Aurora-A haveto develop in the genetic background with down-regulatedSLAN. To confirm the idea, 10 pairs of lung tissues containingtumor (T) or adjacent normal (N) tissues were subjected toWestern blot adopting anti-SLAN, anti-Aurora-A, or anti-actin antibody (Figure 6A) followed by determination of theband intensity by densitometer (Figures 6B, C). Among the

FIGURE 2. SLAN is associated with spindle matrix-like structure. A, EGFP-SLAN is localized to the vicinity of spindle withoutfiber appearance in H1299, 293T, and HeLa. The three different cells transfected with EGFP-SLAN were subjected to immuno-fluorescence using anti-�-tubulin antibody. DNA was labeled with DAPI. B, The distribution of SLAN is sensitive to nocodazoleor Triton X-100. H1299 cells harboring EGFP-SLAN were challenged with 1 �g/ml nocodazole for 8 minutes or 0.01% TritonX-100 for 10 minutes and then applied to immunofluorescence adopting anti-�-tubulin antibody. DNA was labeled withDAPI. SLAN, suppressed in lung cancer; EGFP, enhanced yellow-green variant of green fluorescent protein; DAPI, 4�,6-di-amidino-2-phenylindole.



10 collected lung cancer tissues, Aurora-A was up-regulatedin nine biopsies, among which seven tumor tissues had lowlevel of SLAN, indicating most of the collected biopsies withelevated expression of Aurora-A grew into tumors whenSLAN was reduced.

DISCUSSIONAccording to our analyses, several lines of evidence

suggest a potential inhibitory effect of SLAN on tumorgrowth. First, SLAN mRNA and protein are reduced in lungcancer tissues. Loss of the DNA region covering SLAN genelocus is reported in several different human cancers (Supple-mentary Table, SDC 2, http://links.lww.com/JTO/A71). Sec-ond, overexpression of exogenous SLAN inhibits cell prolif-eration. By contrast, knockdown of SLAN stimulates cell

proliferation, even in low-serum environment. At last, SLANreduces the level of active Aurora-A in cells and directlyinhibits the kinase activity of the latter, an oncoproteinoverexpressed in more than 20 human cancers.

SLAN is a protein with multiple subcellular localizations.SLAN contains an ER membrane retention signal and is colo-calized with EYFP-ER marker. SLAN has two NLS and twonucleus export signals, indicating a complicated regulation ofSLAN shuttling between nucleus and cytoplasm. SLAN2-599 is associated with nucleolus, and the number of nucle-olus is increased when SLAN 2-599 is overexpressed (Sup-plementary Figure 2, SDC 3, http://links.lww.com/JTO/A72).Together with the observation that SLAN contains a ribo-somal L7Ae/Gadd45 signature, which is usually found inribosomal proteins or proteins engaged in transfer RNA

FIGURE 3. SLAN hampers cell proliferation. A and B, SLAN inhibits cell proliferation; 293T or H1299 cells transfected withEGFP or EGFP-SLAN for 36 hours were examined with fluorescent microscope, photographed (A), or applied to cell prolifera-tion assay (B). C, Knockdown of SLAN accelerates cell proliferation; 293T cells transfected with EGFP or EGFP plus each one ofthe four different SLAN shRNAs (namely C, E, F, and G) with the molar ratio 1:10 were subjected to Western blot with anti-SLAN antibody. The excess amount of shRNA may guarantee the expression of SLAN shRNA in EGFP transfected cells. The cellproliferation assay was parallely performed. D, Knockdown of SLAN stimulates cell growth in low serum; 293T cells transfectedwith EGFP or EGFP plus SLAN shRNA-E with the molar ratio 1:10 were subjected to cell proliferation assay. * and *** indicatestatistic significance with p value less than 0.05 and 0.001, respectively, by Student’s t test. SLAN, suppressed in lung cancer;EGFP, enhanced yellow-green variant of green fluorescent protein; shRNA, short hairpin RNA.





processing, ribosomal RNA modification, ribosome biogen-esis, and translation termination,31 all suggest SLAN mighthave a role in ribosome biogenesis or translation. In addition,SLAN is localized to the vicinity of spindle pole with fuzzyappearance, which is akin to spindle matrix.29 Spindle matrixis sensitive to detergent but not nocodazole in Drosophila andXenopus system.29,30 Nevertheless, the distribution of SLANis disrupted on the exposure of mammalian cells to bothnocodazole and detergent, different from the reports aboutthe aspects of spindle matrix in frog and fly. The obser-vation argues that either SLAN is attached to an unknown

structure associated with spindle or the mammalian spindlematrix has distinct property. Moreover, SLAN blockscytokinesis and is found located at midbody at the last stepof mitosis, where a contractile ring is formed to executethe separation of cytoplasm, thereby hinting that SLANmight monitor and prevent the premature execution ofcytokinesis. On the other hand, SLAN 2-551 or 2-599mutant is detected on spindle pole. Overexpression of thetwo mutants triggers the malformation of spindle pole,suggesting SLAN has the potential to bind spindle, whichmight be important to the maintenance of spindle morphol-

FIGURE 4. SLAN regulates mitosis. A, Overexpression of SLAN per se or its shRNA affects the proportion of G2/M; 293T cellstransfected with EGFP, EGFP-SLAN, or SLAN shRNA-E were applied to flow cytometer. The percentage of cells residing in eachphase of a cell cycle was calculated and plotted. B, Time-lapse recording of cells harboring EGFP-SLAN; 293T cells transfectedwith EGFP or EGFP-SLAN were applied to time-lapse microscopic analyses. One and two representatives of EGFP and EGFP-SLAN transfected cells were displayed with time sequence labeled on the top. The arrows indicate two nuclei due to failure ofcytokinesis in the cell. C and D, SLAN increases the percentage of binucleated cells and mitotic cells; 293T cells transfectedwith EGFP or EGFP-SLAN were fixed and stained with DAPI, and the percentage of binucleated cells (C) or mitotic cells (D)were counted. E, SLAN full length and deletion mutants partially arrest cells in mitosis; 293T cells transfected with EGFP orvarious deletion constructs of EGFP-SLAN were stained with DAPI and examined with florescence microscope. The percentageof cells residing in mitosis was counted and plotted. F, SLAN 2-551 disturbs �- and �-tubulin; 293T cells transfected withEGFP-SLAN 2-551 were applied to immunofluorescence with anti-�-tubulin or anti-�-tubulin antibody. *, **, and *** representstatistic significance by Student’s t test with p less than 0.05, 0.01, and 0.001, respectively. SLAN, suppressed in lung cancer; EGFP,enhanced yellow-green variant of green fluorescent protein; DAPI, 4�,6-diamidino-2-phenylindole; shRNA, short hairpin RNA.



ogy. Taken together, all the collected data reveal thatSLAN is a protein potentially with multiple signals guid-ing the protein to several subcellular locations and endow-ing the protein with activities to deal with many aspects ofcell physiology.

Overexpression of SLAN inhibits cell proliferation,leading to the failure of establishing SLAN stable clones inseveral cell lines. Flowcytometric analysis reveals that exog-enous SLAN increases G2/M portion, and time-lapse record-ing shows that the SLAN either arrests cells in mitosis orblocks cytokinesis, which keeps cell number as one after

long-term incubation. Moreover, it is worth pointing out thatthe summation of the percentage of the cells arrested inmitosis and the cells with blocked cytokinesis is almostequivalent to the difference of proliferating cells betweencells overexpressing EGFP-SLAN and EGFP empty vector,indicating the inhibitory effect of SLAN on cell proliferationcan be perfectly explained by the arrest of cells in mitosis andthe blockage of cytokinesis. It remains completely obstacleabout how SLAN regulates mitosis progression and cytoki-nesis. Nevertheless, the findings that SLAN is localized to theproximity of spindle pole and midbody, and SLAN 2-551 or

FIGURE 5. SLAN binds and inactivates Aurora-A. A, Interaction of SLAN with Aurora-A; 293T cells transfected with EGFP-SLAN were treated with nocodazole to enrich mitotic cells and then applied to immunoprecipitation using antibody againstAurora-A (left) or GFP (right). Subsequently, Western blot using antibody against GFP or Aurora-A was conducted. The molec-ular weight of Aurora-A is 46 kD, slightly smaller than Immunoglobulin G heavy chain (IgG-H). B, Aurora-A binds to SLAN�sC-terminal; 293T cells transfected with EGFP-SLAN 2-599 or 474-1101 were exposed to nocodazole and then applied to cellextract preparation and immunoprecipitation with antibody against Aurora-A. Western blot was then followed with antibodyagainst GFP or Aurora-A. C, The kinase activity of Aurora-A is not required for the interaction with SLAN; 293T cells trans-fected with EGFP-SLAN 474-1101 and FLAG-Aurora-A WT or kinase dead (KD) mutant were subjected to immunoprecipitationwith antibody against FLAG and which was followed by Western blot using anti-GFP or FLAG antibody. D, Partial colocaliza-tion of SLAN and Aurora-A; 293T cells cotransfected with EGFP-SLAN and Aurora-A-DsRed were fixed, permeabilized, andstained with DAPI followed by examination with fluorescence microscope. E, Overexpression of exogenous SLAN reduces thelevel of active Aurora-A. EGFP or EGFP-SLAN was expressed in 293T cells followed by cell extract preparation and Western blotusing antibody against SLAN, Aurora-A or phospho-Aurora-A at threonine 288, which represents activated Aurora-A. F, SLANinhibits the activity of Aurora-A. EGFP or EGFP-SLAN was overexpressed in 293T cells followed by isolating EGFP or EGFP-SLANby performing immunoprecipitation using antibody against GFP. The purified EGFP or EGFP-SLAN was added to an in vitrokinase reaction buffer containing recombinant Aurora-A, (�-P33)-ATP, and an artificial substrate MBP. The resulted sampleswere applied to SDS-PAGE, transblotting, and autoradiography or Western blot using anti-GFP antibody. Sup, supernatants;ppt, pellets; MBP, myelin basic protein, SLAN, suppressed in lung cancer; EGFP, enhanced yellow-green variant of green fluo-rescent protein; DAPI, 4�,6-diamidino-2-phenylindole; SDS-PAGE, sodium dodecyl sulfate polyacrylamide gel electrophoresis;WT, wild type; GFP, green fluorescent protein; ATP, adenosine triphosphate; MBP, myelin basic protein.



2-599 is associated with spindle pole and able to disturbtubulin may provide clues to the future exploration of thecorresponding issues.

SLAN binds and inactivates Aurora-A. Aurora-A is anoncokinase transmitting oncogenic signaling through its func-tional kinase activity21 and in turn relaying that to its down-stream substrates.32 Therefore, inactivation of Aurora-A be-comes a powerful strategy to the anticancer therapy. Indeed,Aurora-A inhibitors are developed extensively and have sig-nificant effects on preventing cancer growth in vitro, animalmodels, and clinical trials.11,15,33–37 SLAN diminishes theactivity of Aurora-A by the following observations: first,SLAN lowers the amount of active Aurora-A in cells; second,SLAN blocks autophosphorylation of Aurora-A in in vitrokinase reactions, which is an indicator for the judgment ofkinase activity; and third, SLAN abolishes Aurora-A-cata-lyzed phosphate incorporation into the artificial substratemyelin basic protein. Moreover, when the scenario comes toclinical biopsies, the coincidence of simultaneous up-regula-tion of Aurora-A and down-regulation of SLAN is frequentlydetected in collected lung cancer tissues, which principallyallows cancer tissues develop vigorously because Aurora-A

can have better activity in absence of SLAN. It is, therefore,extremely important to correlate the cancer stages or patientsurvival with the expression pattern of Aurora-A and SLANin a large number of biopsies. Collectively, Aurora-A is awell-known oncoprotein, and SLAN acts similar to a poten-tial tumor suppressor. The inhibitory effect of SLAN on theactivity of Aurora-A could provide a new molecular approachto deal with Aurora-A-dependent cancer incidences.

ACKNOWLEDGMENTSSupported by grants from Taichung Veterans General

Hospital in Taiwan (TCVGH-TCVGH-994701B and TCVGH-TCVGH-994704D).

The authors thank Mei-Chun Liu in Instrument Centerof Taichung Veterans General Hospital for technical support.

REFERENCES1. Yeang CH, McCormick F, Levine A. Combinatorial patterns of somatic

gene mutations in cancer. FASEB J 2008;22:2605–2622.2. Simpson AJ. Sequence-based advances in the definition of cancer-

associated gene mutations. Curr Opin Oncol 2009;21:47–52.3. Alberg AJ, Brock MV, Samet JM. Epidemiology of lung cancer: looking

to the future. J Clin Oncol 2005;23:3175–3185.

N T N T N T826 827 839 3

mal]

N T N T N T

WB: anti-SLAN

WB: anti-Aurora-A

WB ti A ti 1.5

2

2.5

[Aur

ora-

Ano

rm

WB: anti-Actin

WB: anti-SLAN

N T N T N T897 905 906

0.5

1

Aur

ora-

Atu

mor

]

1.2

1.4

al]

WB: anti SLAN

WB: anti-Aurora-A

WB: anti-Actin

0

826 827 839 897 905 906 916 936 944 947

[A

Patients

0.6

0.8

1

1.2

r] [S

LAN

norm

a

WB: anti-SLAN

916 936 944 947N T N T N T N T

0

0.2

0.4

[SLA

Ntu

mor

WB: anti-Aurora-A

WB: anti-Actin

826 827 839 897 905 906 916 936 944 947

Patients

A B

C

FIGURE 6. The protein expression of SLAN or Aurora-A in 10 lung cancer tissues. A, The biopsies from paired lung tumor (T)or adjacent normal tissues (N) were applied to Western blot using antibody against SLAN, Aurora-A, or actin. The numbersabove represent the codes of patients. Actin served as a loading control. B and C, Quantification of SLAN or Aurora-A wasdone by measurement of the protein level by densitometer. The protein level of Aurora-A or SLAN was normalized againstthat of actin. Sequentially, the ratio of normalized Aurora-A (B) or SLAN (C) in tumor over that in normal tissue was calculatedand plotted. SLAN, suppressed in lung cancer.



4. Youlden DR, Cramb SM, Baade PD. The international epidemiology oflung cancer: geographical distribution and secular trends. J ThoracOncol 2008;3:819–831.

5. Ewart-Toland A, Dai Q, Gao YT, et al. Aurora-A/STK15 T�91A is ageneral low penetrance cancer susceptibility gene: a meta-analysis ofmultiple cancer types. Carcinogenesis 2005;26:1368–1373.

6. Gu J, Gong Y, Huang M, et al. Polymorphisms of STK15 (Aurora-A)gene and lung cancer risk in Caucasians. Carcinogenesis 2007;28:350–355.

7. Ogawa E, Takenaka K, Katakura H, et al. Perimembrane Aurora-Aexpression is a significant prognostic factor in correlation with prolifer-ative activity in non-small-cell lung cancer (NSCLC). Ann Surg Oncol2008;15:547–554.

8. Agnese V, Bazan V, Fiorentino FP, et al. The role of Aurora-A inhibitorsin cancer therapy. Ann Oncol 2007;6:vi47–vi52.

9. Boss DS, Beijnen JH, Schellens JH. Clinical experience with aurorakinase inhibitors: a review. Oncologist 2009;14:780–793.

10. Carvajal RD, Tse A, Schwartz GK. Aurora kinases: new targets forcancer therapy. Clin Cancer Res 2006;12:6869–6875.

11. Cheung CH, Coumar MS, Hsieh HP, et al. Aurora kinase inhibitors inpreclinical and clinical testing. Expert Opin Investig Drugs 2009;18:379–398.

12. Coumar MS, Cheung CH, Chang JY, et al. Advances in Aurora kinaseinhibitor patents. Expert Opin Ther Pat 2009;19:321–356.

13. Gautschi O, Heighway J, Mack PC, et al. Aurora kinases as anticancerdrug targets. Clin Cancer Res 2008;14:1639–1648.

14. Kitzen JJ, de Jonge MJ, Verweij J. Aurora kinase inhibitors. Crit RevOncol Hematol 2010;73:99–110.

15. Moore AS, Blagg J, Linardopoulos S, et al. Aurora kinase inhibitors:novel small molecules with promising activity in acute myeloid andPhiladelphia-positive leukemias. Leukemia 2010;24:671–678.

16. Copeland PR, Driscoll DM. RNA binding proteins and selenocysteine.Biofactors 2001;14:11–16.

17. Mudge J, Miller NA, Khrebtukova I, et al. Genomic convergenceanalysis of schizophrenia: mRNA sequencing reveals altered synapticvesicular transport in post-mortem cerebellum. PLoS One 2008;3:e3625.

18. Hawse JR, Hejtmancik JF, Huang Q, et al. Identification and functionalclustering of global gene expression differences between human age-related cataract and clear lenses. Mol Vis 2003;9:515–537.

19. Cahill S, Smyth P, Finn SP, et al. Effect of ret/PTC 1 rearrangement ontranscription and post-transcriptional regulation in a papillary thyroidcarcinoma model. Mol Cancer 2006;5:70.

20. Yu JX, Sieuwerts AM, Zhang Y, et al. Pathway analysis of genesignatures predicting metastasis of node-negative primary breast cancer.BMC Cancer 2007;7:182.

21. Bischoff JR, Anderson L, Zhu Y, et al. A homologue of Drosophila

aurora kinase is oncogenic and amplified in human colorectal cancers.EMBO J 1998;17:3052–3065.

22. Zhou H, Kuang J, Zhong L, et al. Tumour amplified kinase STK15/BTAK induces centrosome amplification, aneuploidy and transforma-tion. Nat Genet 1998;20:189–193.

23. Wang X, Zhou YX, Qiao W, et al. Overexpression of aurora kinase A inmouse mammary epithelium induces genetic instability preceding mam-mary tumor formation. Oncogene 2006;25:7148–7158.

24. Crane R, Gadea B, Littlepage L, et al. Aurora A, meiosis and mitosis.Biol Cell 2004;96:215–229.

25. Marumoto T, Zhang D, Saya H. Aurora-A—a guardian of poles. Nat RevCancer 2005;5:42–50.

26. Barr AR, Gergely F. Aurora-A: the maker and breaker of spindle poles.J Cell Sci 2007;120:2987–2996.

27. Su LJ, Chang CW, Wu YC, et al. Selection of DDX5 as a novel internalcontrol for Q-RT-PCR from microarray data using a block bootstrapre-sampling scheme. BMC Genomics 2007;8:140.

28. Lin YS, Su LJ, Yu CT, et al. Gene expression profiles of the aurorafamily kinases. Gene Expr 2006;13:15–26.

29. Tsai MY, Wang S, Heidinger JM, et al. A mitotic lamin B matrixinduced by RanGTP required for spindle assembly. Science 2006;311:1887–1893.

30. Walker DL, Wang D, Jin Y, et al. Skeletor, a novel chromosomal proteinthat redistributes during mitosis provides evidence for the formation ofa spindle matrix. J Cell Biol 2000;151:1401–1412.

31. Caban K, Copeland PR. Size matters: a view of selenocysteine incor-poration from the ribosome. Cell Mol Life Sci 2006;63:73–81.

32. Yu CT, Hsu JM, Lee YC, et al. Phosphorylation and stabilization ofHURP by Aurora-A: implication of HURP as a transforming target ofAurora-A. Mol Cell Biol 2005;25:5789–5800.

33. Arbitrario JP, Belmont BJ, Evanchik MJ, et al. SNS-314, a pan-Aurorakinase inhibitor, shows potent anti-tumor activity and dosing flexibilityin vivo. Cancer Chemother Pharmacol 2010;65:707–717.

34. Benten D, Keller G, Quaas A, et al. Aurora kinase inhibitor PHA-739358 suppresses growth of hepatocellular carcinoma in vitro and in axenograft mouse model. Neoplasia 2009;11:934–944.

35. Lin ZZ, Hsu HC, Hsu CH, et al. The Aurora kinase inhibitor VE-465 hasanticancer effects in pre-clinical studies of human hepatocellular carci-noma. J Hepatol 2009;50:518–527.

36. Ochi T, Fujiwara H, Yasukawa M. Aurora-A kinase: a novel target bothfor cellular immunotherapy and molecular target therapy against humanleukemia. Expert Opin Ther Targets 2009;13:1399–1410.

37. VanderPorten EC, Taverna P, Hogan JN, et al. The Aurora kinaseinhibitor SNS-314 shows broad therapeutic potential with chemothera-peutics and synergy with microtubule-targeted agents in a colon carci-noma model. Mol Cancer Ther 2009;8:930–939.



the novel protein suppressed in lung cancer down … · chang-tze ricky yu, phd,*† jiun-yi hsia,...

Documents